The designs encoded by natural systems are sometimes optimized by evolution to provide solutions to the physical problems posed by both (i) individual development and (ii) continued viability across many environments and evolution. However, our models of natural systems, at the molecular and cellular scales, are largely descriptive. Our models typically recapitulate what parts are involved in a process, when and where a process occurs, and how the process works. Constraints due to why a system might be designed in a particular way are typically missing. We also often lack precise experimental data describing the physical state of our system over time and, when we attempt to model the dynamic behavior of the system, make use of a modeling framework that is based on a hard-sphere dilute gas approximation for the inside of a cell. Not surprisingly, as our models increase in size and complexity, their utility in helping us to predict the behavior that results from novel perturbations to molecular and cellular systems is limited. What is the best practical foundation on which to ground models for the dynamic behavior of many- component molecular and cellular systems? Can we better constrain our models of natural biological systems by exploring why the systems might be designed as we find them? Do our experiments produce enough understanding of the information encoded in natural genetic systems such that we should now expect to be able to predict their behavior? Could we replace natural living systems with engineered surrogates that are better defined and easier to model, interact with, and predict? To make progress on these questions we plan to work with a relatively well-characterized natural biological system, bacteriophage T7. Specifically, we propose to: (1) Design and construct a synthetic bacteriophage T7 genome that is based on our current understanding of the genetic elements that contribute to phage gene expression, (2) Test our models for the parts encoded on the T7 genome by characterizing the behavior of our synthetic genome relative to the behavior of the original 'wild-type1 isolate, (3) Constrain our models for how the parts of T7 produce a functioning whole by testing the hypothesis that feedback control during early and middle T7 gene expression is used to tune the allocation of expression resources across a range of uncertain cellular environments.